Alzheimer’s disease and Aβ

AD is the most common form of dementia and affects approximately one tenth of the elderly pop-ulation above 65 years of age (Gaugler et al., 2019). The formation of Aβ fibrils and their accu-mulation as senile plaques constitute one of the two major molecular hallmarks of AD, the other one being the presence of intracellular tau tangles. Since the first case study by Alois Alzheimer, evidence that favors the key role of Aβ in AD has grown drastically, however, the exact relation-ship between Aβ deposits, tau tangles and AD-associated cognitive decline is still not established (Alzheimer, 1907). Over the years several ideas have been presented for the placement of Aβ-induced neurotoxicity and other key features of AD, some of which are stated as follows (Du et al., 2018; Kinney et al., 2018; Pardo, 2019);

1. Amyloid cascade hypothesis: This hypothesis is one of the earliest ones explaining the pathophysiology of AD and states that mismetabolism of Aβ and the subsequent fibril for-mation initiates AD.

2. Oligomer hypothesis: Primarily an extension of amyloid cascade hypothesis, this hypoth-esis states that oligomeric species, instead of fibrils, are the primary culprits behind AD.

3. Tau hypothesis: In comparison to its former counterparts, this school of thought focuses on the second molecular hallmark of AD, the tau tangles, and states that tau pathology precedes Aβ deposition and causes AD.

4. Inflammation hypothesis: According to this hypothesis, aberrant activation of microglia-associated pathways modulates Aβ and tau pathology and drives AD.

5. Oxidative stress hypothesis: This hypothesis acknowledges Aβ-induced mitochondrial dysfunction and oxidative stress as the cause of AD,

14 6. Metabolic syndrome hypothesis: This idea suggests that AD is a product of

age-associ-ated aberrations in cerebral glucose metabolism and leads to deposition of Aβ.

Nevertheless, every hypothesis acknowledges the involvement of Aβ in AD, either as a cause or consequence of underlying pathology, due to several known facts. Firstly, mutations in APP and Aβ-processing enzymes, PSEN1 and PSEN2, are major causes of the familial variant of this dis-ease (Goate et al., 1991; Haass, 1996; Plassman and Breitner, 1996). Similarly, genetic interven-tions to mutate these genes cause AD-like pathology in experimental models (Kitazawa et al., 2012). Moreover, directly injecting brain-derived Aβ in rodents also leads to neurodegeneration (Ruiz-Riquelme et al., 2018). Lastly, clinical studies show that the presence of Aβ plaques in fron-toparietal regions of the brain precedes tau pathology and cognitive symptoms of the disease, in-dicating its pivotal role in disease pathology (Figure 4).

Figure 4: Proposed timeline of AD-associated changes in the brain. Aberrations in CSF Aβ levels and appearance of plaques precede tau pathology, brain atrophy and cognitive symptoms, indicating the pivotal role of Aβ pathology in AD (Stanley et al., 2016).

However, the repeated failures of Aβ-targeting drugs and the presence of Aβ deposits in non-demented individuals question Aβ-related hypotheses of AD and give some evidence in the favor of other hypotheses (Rodrigue et al., 2009; Du et al., 2018). The identification and isolation of clinically relevant Aβ proteoforms and conformers is therefore necessary.

1.2.1 Clinical features of classical AD

Clinically, AD is defined as memory impairment accompanied by changes in executive function, visuospatial capability, speech, behavior and/or movement. Although a definite diagnosis is still not possible before the autopsy, the following criterion is utilized for diagnosis of probable AD (Schmidt et al., 2012; Jack et al., 2018; Baiardi et al., 2019):

15 1. Decline of three Mini-Mental State Examination (MMSE) points/year

2. Increased tau/phospho-tau (p-tau) and decreased Aβ42 levels in CSF 3. Reduced hippocampal volume

4. Hypometabolism in the parietal lobe, temporal lobe and hippocampus 5. Positive amyloid positron-emission tomography

The patients with early AD present problems with recent episodic memory followed by the devel-opment of progressive anomia. Aphasia is the next symptom to be reported in most cases along with dysexecutive syndrome. Psychiatric symptoms, including irritability, delusions and halluci-nations, are also reported. In the final stages, the patient loses mobility and death occurs due to complications associated with the aforementioned symptoms. The patients survive between 8 to 10 years from the onset of symptoms, as currently there is no cure available for AD (Tang-Wei et al., 2005). The symptoms are managed by acetylcholine esterase inhibitors and memantine (Shao, 2015).

1.2.2 Clinical variants of Alzheimer’s disease

AD is a complex disease that features several different clinical variants based on the age of onset, pathological burden, cognitive decline and psychiatric symptoms, some of which are discussed as follows.

1.2.2.1 Familial AD

Although the age of onset in most cases is around 65 years, onset has been observed in a small fraction of patients (1%) as early as 46 years. These cases generally have the familial or autosomal dominant variant of AD (fAD) with mutations in APP, PSEN1, PSEN2 or one of the other 31 risk genes (Moustafa et al., 2017). Heterogeneity within this variant arises from the differential presen-tation of cognitive symptoms in cases with different mupresen-tations (Ryan et al., 2016).

1.2.2.2 Sporadic AD

Early-onset AD (EOAD) has also been observed without genetic causes and constitutes 5% of all AD cases. However, most cases present late-onset AD (LOAD). Both EOAD and LOAD occur due to sporadic causes, but diabetes mellitus, obesity, smoking, lack of activity and ApoE genotype are thought to act as risk factors (Toyota et al., 2007; Awada, 2015; Crous-Bou et al., 2017).

16

1.2.2.3 Atypical variants of AD

Depending on the affected brain regions, AD can feature an atypical combination of symptoms.

Posterior cortical atrophy is frequently associated with AD pathology in visual association areas and presents worse visual deficits. Similarly, primary progressive aphasia features AD pathology in conjunction with language impairment. Aβ deposits and tau pathology are also common in pa-tients of other neurodegenerative diseases like Parkinson’s disease (PD), Creutzfeldt–Jakob dis-ease (CJD), Gerstmann-Sträussler Syndrome (GSS), dementia with Lewy bodies (DLB), and fron-totemporal dementia (FTD; Mastaglia et al., 1989, Haltia et al., 1991; Amano et al., 1992, Bar-cikowska et al., 1995, Forman et al., 2006).

1.2.2.4 Rapidly progressive dementia with AD pathology

Rapidly progressive dementias constitute a small subset of dementia patients that are characterized by reports of dementia within 1-2 years (weeks in some cases) of disease onset. The short duration of the disease gives an even shorter window for accurate diagnosis and treatment, presenting a challenge for neurologists and biomedical researchers alike. However, if diagnosed in time, many cases are treatable. The most common causes of rapid progression include vascular anomalies, infections, toxic-metabolic causes, autoimmune diseases, metastasis, iatrogenic causes, neuro-degenerative disorders and seizures (Paterson et al., 2012). Although the exact contribution of each of these causes towards the incidence of rapidly progressive dementias is variable in reports from different centers, most cases are attributed to autoimmune diseases and neurodegenerative pathol-ogies. Within the latter untreatable cause, prion diseases, AD and FTD are the most common con-tributing pathologies. Corticobasal syndrome and DLB also contribute towards rapidly progressive dementias (Poser et al., 1999; Papageorgiou et al., 2009; Neto et al., 2017; Geut et al., 2019).

Owing to its contribution towards the etiology of rapidly progressive dementia, rpAD has now been recognized as an atypical clinical variant of AD. The first paper about rpAD was published in 1989, followed by other reports where AD was misdiagnosed as CJD due to rapid deterioration in memory and shorter survival time (Mann et al., 1989; Poser et al., 1999; Reinwald et al., 2004).

Although rpAD researchers and neurologists have not reached a consensus regarding the clinical definition of this disease, many use a decline of at least 6 MMSE points per year and disease duration of less than 3 years (2 years in some centers) as a diagnostic criterion (Figure 5; Abu-Rumeileh et al., 2018; Pillai et al., 2018).

17 Figure 5: Differences among sAD and rpAD. rpAD follows the same clinical course as classical AD, hereafter referred to as sAD, but the progression is faster and the survival is usually less than three years from the onset of symptoms.

1.2.2.4.1 Clinical and molecular differences in sAD and rpAD

Several differences have been observed in clinical course and biomarker profiles among sAD and rpAD cases. Neurological signs, including executive dysfunction, language impairment and move-ment disorder, are observed earlier during the disease course in rpAD cases. Moreover, these cases show higher levels of tau and p-tau along with reduced Aβ42 in CSF in comparison to sAD, how-ever, the utility of these biomarkers to differentiate sAD from rpAD is still debatable (Llorens et al., 2016). 14-3-3, on the other hand, is only present in rpAD cases and can be used for differential diagnosis (Schmidt et al., 2010; Schmidt et al., 2012, Karch et al., 2016). On an anatomic level, no significant differences are observable in brain atrophy and hippocampal volume. In the context of risk factors, APOE ε4 allelic frequency appears to be lower in rpAD cases in comparison to sAD (Ba et al., 2017; Pillai et al., 2018).

The molecular mechanisms behind rapid progression observed in rpAD are yet to be elucidated.

Markers for inflammation (cartilage glycoprotein YKL-40), tissue damage (α-synuclein) and ax-onal damage (neurofilament light) show no significant differences among sAD and rpAD cases.

Moreover, no differences in distribution and structures of plaques and NFTs have been reported (Schmidt et al., 2012). Rapid progression has been attributed to higher levels of PrP^C, a known Aβ receptor, although its levels are also not significantly different among the two variants (Abu-Ru-meileh et al., 2018). However, the presence of different structures and interactors of PrP^C have

18 been validated in rpAD (Zafar et al., 2017). On the proteomic level, plaques in rpAD have several proteins associated with synaptic dysfunction along with fewer active plaque-clearing astrocytes (Drummond et al., 2017).